We want to understand functions that are distinctive to human tumor viruses in pre-neoplastic lesions and in tumor cells. These functions are potential targets for specific anti-viral therapies to treat viral associated cancers. We (Drs. Lambert and Sugden) have studied the replication of EBV and HPV for as long as their plasmid replicons have been known and recently extended our work to include KSHV. The goals of our research have been to characterize in detail the synthesis and partitioning of these viral replicons both to understand them in general and to uncover their unique features as targets for anti-viral, anti-tumor therapies. During this funding period we have developed a method to visualize individually EBV's plasmid replicons in live cells throughout a cell cycle. This approach has revealed a non-random, efficient mechanism for EBV's partitioning and an intrinsic inefficiency in its DNA synthesis. We shall extend this approach to study the synthesis and partitioning of intact genomes of EBV, KSHV, and HPV16 in their natural host cells. In particular, we shall characterize the replication of EBV during the early steps of its infection of primary B-cells to identify its mechanism of establishment, examine the efficiencies of synthesis and partitioning of KSHV as a function of the number of its terminal repeats, and determine the efficiencies and modes of the synthesis and of partitioning of HPV16 in two kinds of epithelial cells. These experiments will help to define characteristics of the synthesis and partitioning of these human tumor viruses which are likely distinct from the human genome and thus potential targets for therapeutic intervention.